RE: criteria for implementing BSV-V2 and BSV-Q
Dear Carolien,
What type of sampling/how many subjects do you have data for?
I think there are three main ways of handling your issue - first you can remove
the the BSV on V2 and or Q (you may only try it on one). The second is to use
the resulting model but to keep in mind that you are dealing with certain
baised diagnostic plots and to interpret carefully. Lastly, if you have prior
data on the compound from other studies you can fix those values to published
or prior results.
I think each approach has its place, and would just keep in mind what you are
trying to use your model for and be careful to document the assumptions you are
making as you make inferences/decisions based on the results from the model.
Best of luck,
Devin Pastoor
Clinical Research Scientist
Center for Translational Medicine
University of Maryland, School of Pharmacy
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of H.C.A.M. Hazendonk
Sent: Wednesday, March 19, 2014 12:54 PM
To: [email protected]
Cc: H.C.A.M. Hazendonk
Subject: [NMusers] criteria for implementing BSV-V2 and BSV-Q
Dear all,
In the present project a two compartmental structural model adequately
describes our data. Estimated parameters are Cl, V1, Q and V2, between-subject
varibility is estimated for CL and V1 - estimates are precise. In the further
development of the structural model introduction of BSV on either V2 and Q
produced a significant drop in objective
function of -17 points. Estimates were however large (>100%) with moderate
precision (40-50%) and large shrinkage (40-50%). We are aware that interpreting
the goodness of fit plots with these shrinkage values is not reliable (Savic &
Karlsson 2009). We are wondering what criteria to use for implementing BSV-V2
and BSV-Q in the structural model.
Kind regards,
Carolien Hazendonk
Drs. H.C.A.M. Hazendonk, M.D. PhD-student
Department of (Pediatric)Hematology,
Erasmus University Medical Center Rotterdam
E-mail address: [email protected]<mailto:[email protected]>